| The type of trade orientation is determined by the degree of openness a country adopts. In strongly outward oriented countries, the degree of openness is the highest. The next highest degree of openness is found in moderately outward oriented countries. Inward oriented countries have a lower degree of openness than outward oriented countries. The lowest degree of openness is found in strongly inward oriented countries. The degree of openness goes up as countries shift from strongly inward to strongly outward oriented strategies. The experience of newly industrializing countries show the superiority of more outward oriented strategies over more inward oriented strategies. More outward oriented countries show a better performance in exports, economic growth, and industrial development. This tendency convinced many developing countries that a more outward oriented trade strategy is more beneficial than a more inward oriented trade strategy. It is believed that the superior performance of an outward oriented trade strategy is not limited to economic growth, but also extends to a better income distribution. From the perspective of the Stolper-Samuelson theorem, when labor abundant countries implement a more outward oriented trade strategy, the share of labor in total income will rise. Also when tariff protection is imposed on imports, the production of importables is expected to increase. In labor abundant countries, the production of importables typically does not use labor intensive production technologies. Therefore, the reward to the abundant factor will decrease, and instead the reward to the scarce factor will increase, as a result of tariff protection. Thus income distribution will change in favor of the scarce factor in response to tariff protection. In other words, freer trade, which is associated with outward orientation, will change income distribution in favor of the abundant factor. The Stolper-Samuelson theorem basically explains the effect of trade policy on functional income distribution. This explanation is limited to how changes in trade policy will affect changes in income distribution between capital and labor. In itself, this explanation does not cover how changes in trade policy will affect income distribution between upper and lower classes in the society. The belief that inequality will decrease with more outward oriented trade strategies is mainly based on the experience of newly industrializing countries. These countries are implementing outward oriented trade strategies which are accompanied by an improvement in personal income distribution. While empirical evidence explaining the relationship between trade orientation and personal income distribution is still scarce, the interest of developing countries in shifting their trade strategy from inward orientation to outward orientation is growing. This study has analyzed the relationship between trade orientation and income distribution in a cross-section of 34 developing countries. By using income distribution data for the 1980s from those countries, this study has tested the hypothesis that an increase in outward orientation in the economy will improve income distribution. In addition to trade policy variables, this study has also examined the effect of income per capita, investment rate, and educational variables on income distribution. The results of this study show that the relationship between income per capita and inequality tend to resemble the pattern established in Kuznets’ relationship. Income per capita shows a positive effect on inequality at the initial stage of development. The relationship between income per capita and inequality turns negative at later stages of development. Initially, there is a tendency for the income share of the bottom 40 per cent to drop as income per capita increases. This negative relationship continues up to a certain level of income per capita. When this level of income per capita is reached, there is a positive relationship between income share of the bottom 40 per cent and income per capita. However, the relationship is not statistically significant. The use of Gini coefficient as a dependent variable also shows the pattern established by Kuznets which in this case is also not statistically significant. This study also found that the investment rate is associated positively with the upper income classes’income share. An increased investment rate is associated with an increased share of the upper income classes. The effect of investment rate on lower income classes is the opposite. An increased investment rate is accompanied by a reduced share of the lower income classes. This negative relationship between investment rate and equality reflects the underlying relationship between saving and income inequality. The upper income classes tends to save more than the lower income classes. Higher income inequality is favorable for higher saving rate. Because the investment rate is influenced by the saving rate, the negative relationship between income distribution and investment rate results from high income inequality leading to high investment. The negative relationship between investment rate and equality might also reflect the effect of labor saving technology introduced in the developing economies. An increase in investment due to the introduction of labor saving technology tend to reduce the demand for labor. This will increase the marginal productivity of labor. Increased marginal productivity of labor due to increased investment goes to the owner of capital rather than goes to labor. Saving out of capital income tends to be higher than saving out of labor income. This tendency deteriorates personal income distribution rather than improves it. The effects of the educational variables available on income distribution are ambiguous. The secondary education enrollment ratio shows a positive association with the share of lower income classes and a negative association with that of upper income classes. Increased secondary school enrollment ratio tends to reduce inequality. The relationship between secondary school enrollment ratio and inequality is statistically significant. This relationship is consistent with the results reported by other studies (Papanek and Kyn, 1984; Bourguignon and Morrison 1990). In contrast, the mean years of schooling are found to have a positive effect on income inequality in the sample. Increased mean year of schooling raises the share of the upper income classes. At the same time, it also reduces the share of the lower income classes. The negative relationship between income distribution and mean years of schooling may reflect the differential access between the upper income classes and the lower income classes to get benefit from higher levels of education. Generally, the upper income classes respond faster to increased opportunity for higher education than the lower income classes. Usually increased opportunity for higher education occurs at the expense of primary and secondary education. This study has attempted to analyze the relationship between trade orientation and income distribution. The results show that outward orientation is positively associated with equality in income distribution. An increase in the degree of outward orientation will reduce income inequality by increasing the share of the lower income classes relative to the share of the higher income classes. It is found that the income share of the bottom 20 per cent and of the bottom 40 per cent of the population increases in response to a reduction in real exchange rate distortion. In other words, an increase in the degree of outward orientation will raise income share of the bottom 20 per cent and of the bottom 40 per cent. To the contrary, the income share of the top 20 per cent of the population decreases as countries become more outward-oriented. Income inequality drops when countries shift their trade strategy from an inward-oriented to an outward-oriented one. This equalizing effect of outward orientation is statistically significant. These findings provide strong support to the existing evidence which showed that income distribution is better at lower levels of trade protection (Bourguignon and Morrison, 1990). The equalizing effect of outward orientation is also found when individual dummy variables for trade orientation are introduced into the model. The analysis shows that inequality is the lowest in countries adopting a strongly outward oriented strategy. In countries adopting a strongly inward oriented strategy, inequality is the highest. Inequality in countries adopting a moderately inward oriented strategy is lower than in countries adopting a moderately outward oriented strategy. Inequality in countries adopting a moderate orientation is higher than inequality in countries adopting a strongly outward oriented strategy, but it is lower than inequality in strongly inward oriented countries. However, it is only the effect of strongly outward oriented strategy that is statistically significant. The use of two stage regression analysis does not change the conclusion derived from the ordinary regression results. An equalizing effect of outward oriented strategy is maintained. The income share of the bottom 20 per cent of the population increases in response to a decrease in real exchange rate distortion or an increase in the degree of openness. The response is statistically significant at the 10 per cent level. Similarly, the second 20 per cent of the population responds positively to a reduced real exchange rate distortion. The response of the second 20 per cent of the population is statistically significant at the 5 per cent level. The real exchange rate distortion does not show a significant effect on income share of the top 20 per cent of the population. A similar result is also noticed from the effect of real exchange rate distortion on income share of the third 20 per cent and of the fourth 20 per cent of the population. However, these results still suggest that lower distortion in real exchange rates is favorable to equality. In the two stage least squares regression analysis, investment rate, INVRATE, is treated as an endogenous variable. It is found that the effect of investment rate on income inequality is slightly reduced. However, the results still show that investment rate is positively related to the share of the upper income classes. Again, the positive relationship between investment rate and income inequality reflects the hypothesis that income inequality affects the investment rate through the saving rate. In addition to the analysis of a cross-section of developing countries, this study has also examined the relationship between trade variables and income distribution in Indonesia. Initially, the country implemented an outward-looking trade strategy until 1973. The strategy then shifted to an inward-looking one which was in place until 1985. Since 1985 Indonesia has shifted her trade strategy from an inward-looking to an outward-looking one. It is believed that changes in trade strategy have a significant impact on economic growth performance and inequality in income distribution. By using data from a cross-section of 26 provinces in Indonesia, this study has examined the effect of outward oriented trade strategy on inequality. Inequality is defined by expenditure distribution and labor income distribution. The assumption is made that the larger the ratio of exports to output, the more outward oriented is an economy. This study has used the log of total exports to output ratio, and the log of non-oil exports to non-oil income ratio. Outward orientation is measured by the ratio of exports to output. The study sample consists of 26 provinces in Indonesia between 1982 and 1990. Income distribution is measured by the share of income of the bottom 40 per cent, middle 40 per cent, and top 20 per cent of the population, the ratio of the top 20 per cent to the bottom 40 per cent, and the Gini coefficient. Outward orientation, income per capita, employment, and secondary education are used to explain variations in inequality. The effect of income per capita does not resemble the pattern of Kuznets’inverted-U curve relationship. The share of the lower income classes initially increases in response to an increase in income per capita. The relationship between the share of the lower income classes and income per capita turned from positive to negative when income per capita increased further. This pattern of relationship between income per capita and inequality is found when using the SUSENAS data for 1982 and 1984. When the SUSENAS expenditure data for 1990 are used, a similar pattern of relationship is also found. In contrast, a Kuznets type of relationship is found between income per capita and inequality when the SUSENAS labor income data for 1990 are used. However, none of these results is statistically significant. The labor force participation rate shows a tendency to have an equalizing effect. The income share of lower income classes is positively associated with an increase in labor participation rate, while that of the higher income classes is negatively associated with it. The opposite associations between the income classes will result in a reduced level of inequality. This equalizing effect is weak and statistically insignificant before 1985. The effect of labor force participation rate in 1980, LFPR80, on inequality in 1982 and 1984 is not statistically significant. After 1985 the equalizing effect shows a mixed result. The effect of labor force participation rate in 1990, LFPR90, on inequality in expenditure distribution in 1990 is still not significant statistically. However, the relationship between LFPR90 and inequality in labor income distribution in 1990 shows an equalizing effect which is statistically significant. An increase in labor force participation rate leads to a reduction in inequality. The relationship between secondary school education and income distribution mostly shows an equalizing effect. The share of lower income classes will increase, while the share of higher income classes falls, in response to an increase in secondary school education. The equalizing effect of secondary school education in 1971, SSENROL71, is found when the relationship is observed from with labor income distribution in 1982. The effect of secondary school education is anti-equalizing between secondary school education and inequality in the 1984 expenditure distribution. However, none of these results is statistically significant. In contrast, analysis relating secondary school education in 1980, SSENROL80, to labor income distribution in 1990 shows that secondary education has a strong equalizing effect. An increase in SSENROL80 will increase the income share of the bottom 40 per cent of the population. A similar effect is also found on the income share of the middle 40 per cent of the population. On the other hand, an increase in SSENROL80 will reduce the income share of the top 20 per cent of the population. Therefore, an increase in secondary school education will lead to a reduction in inequality. This equalizing effect is statistically significant. The equalizing effect of SSENROL80 is stronger when total exports, LNXYTOT87, is used rather than non-oil exports, LNXYNOIL87, in the regression equations. This result is consistent with the significant role of government expenditure in secondary education which is influenced by the growth in oil revenue. Since oil sectors are owned by the government, oil revenue affects inequality through budget allocations from the central to the provincial governments. Export expansion shows an equalizing effect. An increase in the ratio of exports to regional gross domestic products is positively associated with the income share of the bottom 40 per cent, and the income share of the middle 40 per cent. In contrast, an increase in the ratio of exports to regional domestic products is negatively associated with the income share of the top 20 per cent. A negative relationship is also found between the ratio of exports to regional gross domestic products and the ratio of the top 20 per cent to the bottom 40 per cent. The equalizing effect of export expansion is also supported by a negative relationship between the ratio of exports to regional gross domestic products and Gini coefficients. In addition, the equalizing effect of export expansion is found to be larger and stronger in labor income distribution than in expenditure distribution. There is also a tendency for the equalizing effect of export expansion to increase through time. The equalizing effect of export expansion is found to be larger and stronger for 1990 than for 1982 and 1984. Non-oil export expansion also tends to show a larger and stronger equalizing effect than total export expansion. Developing countries adopt an inward-oriented trade strategy at the lower stage of development. At this stage, an increase in income per capita is associated with higher inequality. When the process of economic development enters a higher stage, developing countries shift from an inward-oriented trade strategy to an outward-oriented trade strategy. An increase in income per capita is associated with lower inequality in countries which adopt an outward-oriented trade strategy. Since lower inequality leads to political stability which is necessary for investment, lower inequality is favorable for economic growth (Perotti, 1992; Alesina and Perotti, 1993; Persson and Tabellini, 1994). Outward orientation is also favorable for economic growth (Leamer, 1988, 1992; Greenway and Nam, 1988; Colombatto, 1992; Edwards, 1992; Dollar, 1992). This study found that outward orientation is positively associated with lower inequality. An outward oriented strategy leads to a positive relationship between income per capita and equality at the lower stage of development, and it strengthens the relationship at the higher stage of development. These results sugggest that a more outward oriented trade strategy seems to be appropriate for developing countries which are attempting to achieve higher economic growth with equity. An outward oriented strategy develops the economy on the basis of comparative advantage. Equity aspects of education policy should be consistent with outward oriented trade strategy. Because more equalized educational advantage will help improve the quality of labor which is the source of comparative advantage in developing countries. The equalizing effect of secondary school enrollment ratio strengthens the necessity to emphasize the equity aspects of the human resource development programs in developing countries. The empirical results of the Indonesian case study also led support to outward oriented strategy and equity in educational programs. Expansion of export, particularly non-oil export expansion, in Indonesia since 1985 is considered as the consequence of outward oriented strategy. The increase in the level of secondary school enrollment ratio since 1980 is certainly the product of education equity in the Indonesia’s human resource development programs. The improved level of education has also contributed to improve Indonesia’s comparative advantage in non-oil exports, particularly manufactured exports. This study is based on cross-section data of income distribution in developing countries. The quality of income distribution data varies across countries. Some data are based on expenditure distribution, and some others are based on income distribution. The time reference of the available data is also diverse. It is almost impossible to pick up sufficient number of sample countries for one time reference, particularly in the period after 1980. In addition, the sample selection of income distribution data is also constrained by the measures for trade orientation which are available only for periods before 1985. Moreover, these measures are not available for every country for years which can match the available data on income distribution. In the case study of Indonesia, the income distribution data is relatively better than the cross countries data. Despite the limited coverage of data for income distribution, the inequality data is available for homogenous time reference and sampling sources. However, like the cross countries data, the Indonesian income distribution data is also available only for particular years across provinces. Although the ratio of exports to output represents the degree of comparative advantage rather than trade orientation, it is the best available device to measure the degree of outward orientation across provinces in Indonesia. A work on the calculation of trade orientation index based on time series data of exchange rate distortion should be started in Indonesia. Further research should focus on the effect of trade orientation on income distribution of individual developing countries through time. This type of study should also cover the impact of trade orientation on functional income distribution, particularly labor with which developing countries abundantly endowed. |
syafruddinkarimi 